Transverse wave prediction method

Abstract

The invention belongs to the technical field of shale oil gas and coal bed gas exploration and development, and particularly relates to a transverse wave prediction method. According to the method, data of transverse wave related parameters of a target well and two binary nonlinear functions corresponding to the transverse wave related parameters and the transverse wave time difference are obtained, and the transverse wave time difference is obtained through calculation and serves as a transverse wave prediction result of the target well; the transverse wave related parameters are longitudinal wave time difference and density; the binary nonlinear function is obtained by dividing the product of the square of the longitudinal wave time difference and the density by a binary linear polynomial formed by the longitudinal wave time difference and the density; therefore, a binary nonlinear function corresponding to the two parameters and the shear wave time difference is established only for the two parameters (the longitudinal wave time difference and the density) with relatively strong correlation with the shear wave, so that the nonlinear mapping form is simpler, the calculation cost can be effectively saved, and meanwhile, due to the fact that the influence of other rock physical parameters with relatively weak correlation is eliminated, the calculation efficiency is improved. And the accuracy of the mapping relation formed by the nonlinear function is ensured.

Description

Technical Field

[0001] This invention belongs to the field of shale oil and gas and coalbed methane exploration and development technology, and specifically relates to a shear wave prediction method. Background Art

[0002] In shale oil and gas exploration, due to the poor physical properties and strong heterogeneity of shale reservoirs, it is necessary to predict sweet spots in tight reservoirs. Shear wave data is a key parameter for predicting sweet spots and evaluating tight reservoirs, and it serves as the foundation for pre-stack seismic data inversion, brittleness factor calculation, stress analysis, and TOC prediction. Accurate P-wave and S-wave logging velocity data are required for pre-stack seismic inversion and pre-stack seismic attribute analysis.

[0003] However, in actual production, due to logging costs and the exploration targets at the time, conventional logging is mostly used, which can only provide P-wave data. Shear wave data is scarce in most work areas. To obtain the necessary shear wave velocity data, many geophysicists have conducted extensive research in this area.

[0004] Currently, most methods rely on acoustic logging data and laboratory test data, using a linear fit to obtain empirical formulas for shear wave velocity. However, the relationship between shear wave velocity and acoustic logging and laboratory test data is not necessarily a simple linear one, resulting in low prediction accuracy for linear fitting methods. Existing technologies also employ multivariate nonlinear shear wave prediction models constructed using rock physical parameters related to shear waves. This significantly improves prediction accuracy through multivariate nonlinear mapping. For example, Chinese invention patent CN116068650A discloses a deep learning-based shear wave velocity prediction method. This method establishes a deep learning model reflecting the relationship between P-wave velocity curves, neutron porosity curves, density curves, porosity curves, and clay content curves and corresponding measured shear wave velocity curves. The corresponding shear wave velocity prediction model is obtained through iterative training. However, constructing such a nonlinear shear wave prediction model typically requires iterative training of a neural network using numerous shear wave-related parameters, a complex process with high computational costs. Summary of the Invention

[0005] The purpose of this invention is to provide a shear wave prediction method to solve the problem that the construction of nonlinear shear wave prediction models in the prior art usually requires iterative training of neural networks with a large number of parameters related to shear waves, which leads to a relatively complex process and high computational cost.

[0006] To achieve the above objectives, this invention provides a shear wave prediction method, which acquires data of two shear wave related parameters of the target well and a bivariate nonlinear function of the two shear wave related parameters with shear wave travel time as the dependent variable, and calculates the shear wave travel time as the shear wave prediction result of the target well; the two shear wave related parameters are P-wave travel time and density;

[0007] The bivariate nonlinear function is obtained by multiplying the square of the longitudinal wave time difference by the density and then dividing by a bivariate linear polynomial composed of the longitudinal wave time difference and the density.

[0008] Furthermore, the bivariate nonlinear function is:

[0009]

[0010] In the formula, DTS is the transverse wave time difference; DEN is the density; DT is the longitudinal wave time difference; and a, b, and c are coefficients in a bivariate linear polynomial composed of the longitudinal wave time difference and the density.

[0011] Furthermore, the coefficients in the bivariate linear polynomial composed of the longitudinal wave time difference and density are obtained by substituting the measured transverse wave, longitudinal wave and density data into the bivariate nonlinear function.

[0012] Furthermore, the shear wave correlation parameters were determined by comparing the correlation between each rock physical parameter and the shear wave time difference.

[0013] Furthermore, the methods for obtaining data on the two shear wave correlation parameters of the target well include:

[0014] Identify the well sections of the target well that require logging data correction, and correct the measured P-wave time difference curve of the well sections; based on the corrected P-wave time difference curve and other well sections of the target well, obtain the density data in the shear wave correlation parameters of the target well.

[0015] Furthermore, the methods for obtaining data on the two shear wave correlation parameters of the target well include:

[0016] The well section of the target well that requires logging data correction is determined, and the measured P-wave time difference curve of the well section is corrected to obtain the P-wave time difference data in the shear wave correlation parameter data of the target well.

[0017] Furthermore, methods for determining well sections requiring logging data correction include:

[0018] If the difference between the well diameter and the drill bit diameter of a certain section of the target well is greater than a set difference threshold, the section is identified as an enlarged section. Based on the overlap between the shale section of the target well and the enlarged section, the section that needs to be corrected for logging data is determined.

[0019] Furthermore, the methods for correcting the measured P-wave time difference curve of the well section include:

[0020] By using the P-wave transit time curve segment of the normal well section, the correlation equation between resistivity and P-wave transit time was constructed:

[0021] dt=kR m

[0022] In the formula, dt is the longitudinal wave time difference, in μs / ft; R is the resistivity, in Ω·m; k and m are coefficients to be calculated.

[0023] By combining the P-wave transit time curve segments and corresponding resistivity curve data of normal well sections, the relationship between the actual P-wave transit time and the predicted P-wave transit time of normal well sections is constructed using the least squares method:

[0024]

[0025] In the formula, Er is the difference between the actual and predicted P-wave transit time in the normal well section, n is the number of sampling points for the P-wave transit time curve and resistivity curve in the normal well section, and DT i Let dt be the actual value of the P-wave time difference at the i-th sampling point. i R is the predicted P-wave time difference for the i-th sampling point. i Let Er be the resistivity value at the i-th sampling point; let Er be the minimum, and solve for the values ​​of the coefficients k and m to be calculated;

[0026] Based on the resistivity data of the well section requiring logging data correction and the correlation equation between resistivity and P-wave travel time, the corrected P-wave travel time curve is obtained.

[0027] Furthermore, the coefficients in the bivariate linear polynomial composed of the longitudinal wave time difference and density are obtained by substituting the measured transverse wave, longitudinal wave, and density data into the bivariate nonlinear function using the Marquardt method.

[0028] The technical solution of this invention improves upon existing methods for predicting shear waves using nonlinear relationships, resulting in the following advantages:

[0029] For only two parameters that are strongly correlated with shear waves (longitudinal wave transit time and density), a bivariate nonlinear function corresponding to the shear wave transit time was established. This simplifies the form of the nonlinear mapping and effectively saves computational costs. At the same time, by excluding the influence of other rock physics parameters with weak correlation, the accuracy of the mapping relationship constructed by the nonlinear function is guaranteed. Attached Figure Description

[0030] Figure 1This is a flowchart of the shear wave prediction method in an embodiment of the present invention.

[0031] Figure 2a This is a schematic diagram illustrating the correlation analysis between density and shear wave time difference in an embodiment of the shear wave prediction method of the present invention;

[0032] Figure 2b This is a schematic diagram illustrating the correlation analysis between P-wave time difference and S-wave time difference in an embodiment of the shear wave prediction method of the present invention;

[0033] Figure 2c This is a schematic diagram illustrating the correlation analysis between gamma and shear wave time difference in an embodiment of the shear wave prediction method of the present invention;

[0034] Figure 2d This is a schematic diagram illustrating the correlation analysis between Poisson's ratio and shear wave time difference in an embodiment of the shear wave prediction method of the present invention;

[0035] Figure 3 This is a schematic diagram of the cross-analysis results of the measured shear wave time difference and the shear wave time difference calculated by the shear wave prediction method in the embodiment of the shear wave prediction method of the present invention.

[0036] Figure 4 This is a schematic diagram comparing the shear wave time difference curve of well W354 obtained by the shear wave prediction method in an embodiment of the present invention with the measured longitudinal and shear wave time difference curves.

[0037] Figure 5 This is a schematic diagram comparing the shear wave velocities obtained from the Xu Huai Te model, the differential equivalent medium model, the self-consistent model, and the shear wave prediction method in the embodiments of the present invention with the measured shear wave velocity curves of the WYHF1 pilot well.

[0038] Figure 6 This is a schematic diagram of the seismic gather analysis of the WYHF1 shale oil pilot well and its adjacent wells in an embodiment of the shear wave prediction method of the present invention;

[0039] Figure 7a This is a schematic diagram of the forward modeling results of the WYHF1 shale oil pilot well and its adjacent wells obtained by performing AVO forward modeling on the prediction results of the shear wave prediction method of the present invention using the prediction results of the shear wave predicted by the Xu Huait model in an embodiment of the shear wave prediction method of the present invention.

[0040] Figure 7b This is a schematic diagram of the AVO characteristics of the WYHF1 shale oil pilot well and its adjacent wells, obtained by performing AVO forward modeling based on the prediction results obtained by predicting shear waves using the Xu Huait model in an embodiment of the shear wave prediction method of the present invention.

[0041] Figure 7cThis is a schematic diagram of the AVO characteristics of the WYHF1 shale oil pilot well and its adjacent wells, obtained by performing AVO forward modeling based on the prediction results obtained by predicting shear waves using the Xu Huait model in an embodiment of the shear wave prediction method of the present invention.

[0042] Figure 7d This is a schematic diagram of the AVO characteristics of the WYHF1 shale oil pilot well and its adjacent wells, obtained by performing AVO forward modeling on the prediction results obtained by predicting shear waves using the Xu Huait model in an embodiment of the shear wave prediction method of the present invention.

[0043] Figure 8a This is a schematic diagram of the forward modeling results of the WYHF1 shale oil pilot well and its adjacent wells obtained by performing AVO forward modeling on the prediction results of the shear wave prediction method in an embodiment of the present invention.

[0044] Figure 8b This is a schematic diagram of the AVO characteristics of the WYHF1 shale oil pilot well and its adjacent wells, obtained by performing AVO forward modeling based on the prediction results obtained by predicting shear waves using the Xu Huait model in an embodiment of the shear wave prediction method of the present invention.

[0045] Figure 8c This is a schematic diagram of the AVO characteristics of the WYHF1 shale oil pilot well and its adjacent wells, obtained by performing AVO forward modeling based on the prediction results obtained by predicting shear waves using the Xu Huait model in an embodiment of the shear wave prediction method of the present invention.

[0046] Figure 8d This is a schematic diagram of the AVO characteristics of the WYHF1 shale oil pilot well and its adjacent wells, obtained by performing AVO forward modeling based on the prediction results obtained by predicting shear waves using the Xu Huait model in an embodiment of the shear wave prediction method of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0048] Examples of Shear Wave Prediction Methods

[0049] This embodiment presents a technical solution for a shear wave prediction method, referring to... Figure 1 The shear wave prediction method obtains the shear wave related parameters of the target well and a bivariate nonlinear function of two shear wave related parameters with shear wave travel time as the dependent variable (i.e., the bivariate nonlinear function has shear wave travel time as the dependent variable and two shear wave related parameters as independent variables), and calculates the shear wave travel time as the shear wave prediction result of the target well; where the two shear wave related parameters are P-wave travel time and density;

[0050] The two shear wave related parameters (i.e., P-wave time difference and density) and the bivariate nonlinear function corresponding to the shear wave time difference are obtained by multiplying the square of the P-wave time difference and the density by a bivariate linear polynomial composed of the P-wave time difference and the density.

[0051] Therefore, the shear wave prediction method in this embodiment only targets two parameters that are strongly correlated with shear waves (longitudinal wave time difference and density), and establishes a binary nonlinear function corresponding to these two parameters and the shear wave time difference. This makes the form of the nonlinear mapping simpler and can effectively save computational costs. At the same time, since the influence of other rock physical parameters with weak correlation is excluded, the accuracy of the mapping relationship formed by the nonlinear function is guaranteed.

[0052] In this embodiment, to ensure higher accuracy of the constructed shear wave prediction model (i.e., the relationship function between P-wave time difference and density and shear wave time difference), thereby improving the accuracy of shear wave prediction, a bivariate fractional shear wave prediction model (i.e., a bivariate nonlinear function) is constructed based on the property that fractional functions have greater approximation and fitting ability than conventional polynomial functions in function approximation theory. Furthermore, based on rock physics correlation analysis, it is known that P-wave velocity has the highest correlation with shear wave velocity, followed by density. Therefore, the P-wave value in the numerator of the bivariate nonlinear function is squared to highlight the influence of P-waves on shear waves. The denominator of the bivariate nonlinear function is in the form of a bivariate polynomial of P-wave and density to improve the adaptability of the bivariate nonlinear function to actual working conditions.

[0053] Specifically, the bivariate nonlinear function is:

[0054]

[0055] In the formula, DTS is the transverse wave time difference, in μs / ft; DEN is the density, in g / cm³. 3 DT represents the P-wave transit time, in μs / ft; a, b, and c are coefficients in a bivariate linear polynomial composed of the P-wave transit time and density. The coefficients a, b, and c in this bivariate linear polynomial are obtained by substituting measured shear wave, P-wave, and density data into the aforementioned bivariate nonlinear function. In this embodiment, the coefficients in the bivariate linear polynomial composed of the P-wave transit time and density are obtained by substituting measured shear wave, P-wave, and density data from normal or corrected well sections into the aforementioned bivariate nonlinear function using the Marquardt method. Since the Marquardt method for solving unknown coefficients is existing technology, it will not be elaborated here.

[0056] The shear wave correlation parameters were determined by comparing the correlation between each rock physical parameter and the shear wave transit time. In other words, in this embodiment, P-wave transit time and density were selected as the shear wave correlation parameters by analyzing the correlation between each rock physical parameter and the shear wave transit time. Since the shear wave velocity (shear wave transit time) of a formation is affected by various factors such as lithology, physical properties, burial depth, density, and fissure fluids, it reflects the unique characteristics and correlations among the rock physical parameters of the formation. Through correlation analysis between shear wave velocity and other rock physical parameters, rock physical parameters with high correlation to shear wave velocity can be identified. (Refer to...) Figures 2a-2d Correlation analysis was conducted on the rock physical parameters of P-wave transit time, density, Poisson's ratio, and gamma obtained from the logging curves of wells W354 and W396 in a certain target area, and the shear wave transit time. It can be seen that the shear wave transit time is highly correlated with the P-wave transit time and density, thus proving the rationality of the shear wave correlation parameters selected in this shear wave prediction method.

[0057] The target area in this embodiment is a Mesozoic-Cenozoic rift basin, containing the Yuhuangding Formation, Dacangfang Formation, Hetaoyuan Formation, and Liaozhuang Formation from bottom to top; among them, the Hetaoyuan Formation is the main exploration target layer. After decades of exploration and development, the proven rate of conventional oil reservoirs in this target area has reached 57%. Because the Hetaoyuan Formation within the depression contains high-quality source rocks, shale oil is the primary exploration target in the near term. However, the mudstone and shale formations contain a large amount of dispersed minerals such as kaolinite and illite, as well as hard and brittle minerals such as quartz, resulting in well-developed fractures and bedding, and high water sensitivity. During drilling, mechanical vibration can easily cause localized shale detachment and collapse, leading to formation alteration and an uneven wellbore, resulting in partial distortion of logging curves. Consequently, the shear wave transit time calculated from the data corresponding to these logging curves is inaccurate. Therefore, in this embodiment, the methods for obtaining the shear wave-related parameters of the target well include:

[0058] The well section of the target well requiring logging data correction is identified. The measured P-wave transit time curve of this section is corrected to obtain the P-wave transit time data from the shear wave correlation parameters of the target well. Based on the corrected P-wave transit time curve, the density data from the shear wave correlation parameters of the target well is obtained. In other embodiments, other methods can also be used to obtain the density data or the P-wave transit time data from the shear wave correlation parameters of the target well.

[0059] After correcting the data for well sections where logging curves may be distorted, density data for calculating shear wave time difference is obtained from the corrected data, which can make the shear wave prediction results more consistent with the actual working conditions.

[0060] The methods for determining the well sections that require logging data correction include:

[0061] If the difference between the well diameter and the drill bit diameter of a certain section of the target well is greater than the set difference threshold, the section is identified as an enlarged section. Based on the overlap between the shale section and the enlarged section of the target well, the section that needs to be corrected for logging data is determined. Specifically, in this embodiment, logging curves, drilling strata, logging data, and core samples from 38 wells in the target area were collected. Since most wells in the target area are old, logging data correction was required. Well sections with diameters greater than drill bit diameters were identified. For the target area in this embodiment (i.e., the area where the target well is located), based on the characteristics of the diameter curve and drill bit diameter, a difference threshold of 0.0432m was set, meaning well sections where the difference between the diameter and drill bit diameter was greater than 0.0432m were classified as enlarged-diameter sections. Furthermore, by statistically analyzing the GR curve and density curve values ​​in the target shale and sandstone layers, the depth of the corresponding shale well section (i.e., the shale well section in the target area) was determined. In this target area, the GR curve value range in the target shale layer is 135-240 API, and the value range in the sandstone layer is 65-133 API; the density curve value range in the target shale layer is 2.4-2.7 g / cm³. 3 The value range for sandstone is 2.2-2.4 g / cm³. 3 Based on the GR and density curves for the target shale and sandstone formations, the depth of the shale section of the target well is determined. The depth of the corresponding well section with a GR value between 135-240 API is obtained from the GR curve, or the depth of the well section with a density value between 2.4-2.7 g / cm³ is obtained from the density curve. 3 The depth of the corresponding well section within the specified range corresponds to the depth of the mudstone and shale strata in that well. Specifically, the depths of the mudstone and shale strata in the target well are 2284m-2897m, 2308m-2321m, 2325m-2330m, 2388m-2399m, 2419m-2430m, 2455m-2484m, and 2492m-2503m.

[0062] After identifying the shale and enlarged sections of the target well, the sections requiring logging data correction are determined based on the overlapping portion of these sections. In this embodiment, the overlapping portion of the shale and enlarged sections is used as the section requiring logging data correction. Therefore, identifying the section requiring logging data correction based on the overlapping portion of the shale and enlarged sections ensures accurate identification of enlarged sections primarily caused by wellbore collapse in the shale sections, minimizing the possibility of misidentifying normal sections as those requiring correction.

[0063] In this embodiment, the method for correcting the measured P-wave time difference curve of the well section requiring logging data correction includes:

[0064] By using the P-wave transit time curve segment of the normal well section, the correlation equation between resistivity and P-wave transit time was constructed:

[0065] dt=kR m

[0066] In the formula, dt is the longitudinal wave time difference, in μs / ft; R is the resistivity, in Ω·m; k and m are coefficients to be calculated.

[0067] By combining the P-wave transit time curve segments and corresponding resistivity curve data of normal well sections, the relationship between the actual P-wave transit time and the predicted P-wave transit time of normal well sections is constructed using the least squares method:

[0068]

[0069] In the formula, Er is the difference between the actual and predicted P-wave transit time in the normal well section, n is the number of sampling points for the P-wave transit time curve and resistivity curve in the normal well section, and DT i Let dt be the actual value of the P-wave time difference at the i-th sampling point. i R is the predicted P-wave time difference for the i-th sampling point. i Let Er be the resistivity value at the i-th sampling point; let Er be the minimum, and solve for the values ​​of the coefficients k and m to be calculated;

[0070] By analyzing the measured P-wave transit time curves of the well sections requiring logging data correction and the correlation equation between resistivity and P-wave transit time, the corrected P-wave transit time curves are obtained. Since resistivity logging has a large detection radius and is less affected by the wellbore environment, the resistivity curves obtained from logging are usually accurate and reliable. Furthermore, the resistivity curves and P-wave transit time curves have a good correlation. Therefore, if the P-wave transit time curves of normal well sections with good wellbore quality are selected, a predictive equation between resistivity and P-wave transit time can be constructed. Then, based on this equation, the P-wave transit time curves of well sections with poor wellbore environments requiring logging data correction can be obtained, leading to more accurate correction results for the P-wave transit time curves.

[0071] For the target area, the above-mentioned method of correcting the measured P-wave transit time curve of the well section requiring logging data correction is applied. The final solution yields the coefficients to be calculated: k = 13.86; m = -0.28. After obtaining the corrected P-wave transit time curve, the density value is calculated using the Gardenr formula, thus obtaining the density data in the target well's shear wave correlation parameters. Specifically, the density curve of the original distorted well section (i.e., the well section requiring logging data correction) can be replaced with the curve constructed from the calculated density values ​​to construct a new density curve. The Gardenr formula is as follows:

[0072] ρ=0.31dt 0.25

[0073] In the formula, ρ is the density value, with units of g / cm³. 3 dt is the longitudinal wave time difference (i.e., the corrected longitudinal wave time difference), in μs / ft.

[0074] Using the shear wave prediction method described above in this embodiment, shear wave prediction is performed on well W354 in the target area; according to... Figure 3 The measured values ​​of well W354 shown (i.e.) Figure 3 The "measured shear wave" in the figure and the shear wave time difference calculated by the shear wave prediction method described above in this embodiment (i.e., Figure 3 According to the cross-plot analysis results of "calculated shear wave" in this embodiment, the correlation coefficient between the shear wave time difference calculated by the shear wave prediction method and the measured shear wave time difference reaches 0.9054; Figure 4 As shown, the shear wave time difference curve of well W354 obtained by the shear wave prediction method of this embodiment is compared with the measured longitudinal and shear wave time difference curves. The consistency is very good. After analysis, the average relative error between the shear wave time difference calculated by the shear wave prediction method of this embodiment and the measured shear wave time difference is less than 4.04%.

[0075] To verify the rationality of the shear wave prediction method in this embodiment, the shear wave velocity curves obtained by the Schwäighter model, the differential equivalent medium model, and the self-consistent model (all commonly used shear wave prediction models) are compared and analyzed with the shear wave velocity curve corresponding to the shear wave time difference calculated by the shear wave prediction method in this embodiment, further determining the effectiveness of this embodiment:

[0076] like Figure 5 As shown, the shear wave velocities obtained by the Xu Huait model, the differential equivalent medium model, the self-consistent model, and the shear wave prediction method of this embodiment are compared with the measured shear wave velocity curves of the WYHF1 guide well. It can be seen that the predicted shear wave curve has the best correlation with the measured value.

[0077] Seismic gather analysis of shale oil was conducted. Through seismic gather analysis of the WYHF1 shale oil pilot well and its adjacent wells, the AVO characteristics of the H33 section were classified as Class IV, exhibiting characteristics such as... Figure 6 The characteristics of energy attenuation from near to far are shown (where Figure 6 The left side shows a schematic diagram of the logging curves of the WYHF1 shale oil pilot well, and the right side shows the pre-stack seismic gathers of the WYHF1 shale oil pilot well and its adjacent wells; it can be seen that the shale sections of multiple wells all exhibit this characteristic.

[0078] Using the Schewoid model to predict shear waves and performing AVO forward modeling, such as... Figures 7a-7d As shown (where Figures 7b-7dThe x-axis represents the incident angle (in °), and the y-axis represents the amplitude. The AVO characteristics of the WYHF1 shale oil pilot well and its adjacent well corresponding to "H33 Zhong-2 Ding" are Class III AVO characteristics, which means that the AVO characteristics obtained from the forward modeling analysis are inconsistent with the actual well-side seismic gather AVO characteristics. After analysis, the reason for this situation may be that the model is a commonly used petrophysical model applicable to sandstone and shale, but not applicable to shale sections, resulting in errors in the predicted shear waves.

[0079] Using the shear wave prediction results from the shear wave prediction method of this embodiment, AVO forward modeling is performed, such as... Figures 8a-8d As shown (where Figures 8b-8d The x-axis represents the incident angle (in °), and the y-axis represents the amplitude. The AVO characteristics of the WYHF1 shale oil pilot well and its adjacent wells in "H33-1 top", "H33-2 top", and "H33-3 top" all exhibit Class IV AVO characteristics, which is consistent with the actual Class IV AVO characteristics. That is, the AVO characteristics of the shale oil section are consistent with the actual well-side seismic gather AVO characteristics, both being Class IV AVO. It can be seen that the shear wave prediction method in this embodiment predicts the shear waves relatively accurately and can be used for pre-stack reservoir prediction.

[0080] This invention has the following characteristics:

[0081] 1) The shear wave prediction method of the present invention only targets two parameters that are strongly correlated with shear waves (longitudinal wave time difference and density), and establishes a binary nonlinear function corresponding to the shear wave time difference for these two parameters, which makes the form of nonlinear mapping simpler and can effectively save computational costs. At the same time, since the influence of other rock physical parameters with weak correlation is excluded, the accuracy of the mapping relationship formed by the nonlinear function is guaranteed.

[0082] 2) The selection of P-wave transit time and density as S-wave correlation parameters was determined by analyzing the correlation between each rock physical parameter and the S-wave transit time; this demonstrates the rationality of the S-wave correlation parameters selected in the S-wave prediction method of this invention.

[0083] 3) First, perform data correction on well sections where logging curves may be distorted, and then obtain density data for calculating shear wave time difference using the corrected data. This will make the shear wave prediction results more consistent with the actual working conditions.

[0084] 4) Based on the overlap between the shale section and the enlarged section, determine the section that needs logging data correction. This ensures that the enlarged section, which is mainly caused by wellbore collapse in the shale section, can be accurately identified, and the situation of misidentifying a normal section as a section that needs correction can be avoided as much as possible.

[0085] 5) Because resistivity logging has a large detection radius and is less affected by the wellbore environment, the resistivity curve obtained from logging is true and reliable. At the same time, the resistivity curve and the P-wave transit time curve have a good correlation. Therefore, this invention selects the P-wave transit time curve of the well section with good wellbore quality, constructs a prediction equation for resistivity and P-wave transit time, and then obtains the P-wave transit time curve of the well section with poor wellbore environment based on this equation, which can obtain a more accurate correction result for the P-wave transit time curve.

[0086] It should be understood that the above-described specific embodiments of the present invention are merely illustrative or explanatory of the principles of the present invention, and do not constitute a limitation thereof.

Claims

1. A method for predicting shear waves, characterized in that, Data on two shear wave correlation parameters of the target well and a bivariate nonlinear function of the two shear wave correlation parameters with shear wave travel time as the dependent variable are obtained. The shear wave travel time is calculated as the shear wave prediction result of the target well. The two shear wave correlation parameters are the longitudinal wave travel time and density. The bivariate nonlinear function is obtained by multiplying the square of the longitudinal wave time difference by the density and then dividing by a bivariate linear polynomial composed of the longitudinal wave time difference and the density.

2. The shear wave prediction method according to claim 1, characterized in that, The binary nonlinear function is: In the formula, DTS is the transverse wave time difference; DEN is the density; DT is the longitudinal wave time difference; and a, b, and c are coefficients in a bivariate linear polynomial composed of the longitudinal wave time difference and the density.

3. The shear wave prediction method according to claim 1 or 2, characterized in that, The coefficients in the bivariate linear polynomial composed of the longitudinal wave time difference and density are obtained by substituting the measured transverse wave, longitudinal wave and density data into the bivariate nonlinear function.

4. The shear wave prediction method according to claim 1 or 2, characterized in that, The shear wave correlation parameters were determined by comparing the correlation between each rock physical parameter and the shear wave time difference.

5. The shear wave prediction method according to claim 1, characterized in that, Methods for obtaining data on the two shear wave correlation parameters of the target well include: Identify the well sections of the target well that require logging data correction, and correct the measured P-wave time difference curve of the well sections; based on the corrected P-wave time difference curve and other well sections of the target well, obtain the density data in the shear wave correlation parameters of the target well.

6. The shear wave prediction method according to claim 1, characterized in that, Methods for obtaining data on the two shear wave correlation parameters of the target well include: The well section of the target well that requires logging data correction is determined, and the measured P-wave time difference curve of the well section is corrected to obtain the P-wave time difference data in the shear wave correlation parameter data of the target well.

7. The shear wave prediction method according to claim 5 or 6, characterized in that, Methods for determining well sections requiring logging data correction include: If the difference between the well diameter and the drill bit diameter of a certain section of the target well is greater than a set difference threshold, the section is identified as an enlarged section. Based on the overlap between the shale section of the target well and the enlarged section, the section that needs to be corrected for logging data is determined.

8. The shear wave prediction method according to claim 5 or 6, characterized in that, The methods for correcting the measured P-wave time difference curve of the well section include: By using the P-wave transit time curve segment of the normal well section, the correlation equation between resistivity and P-wave transit time was constructed: dt=kR m In the formula, dt is the longitudinal wave time difference, in μs / ft; R is the resistivity, in Ω·m; k and m are coefficients to be calculated. By combining the P-wave transit time curve segments and corresponding resistivity curve data of normal well sections, the relationship between the actual P-wave transit time and the predicted P-wave transit time of normal well sections is constructed using the least squares method: In the formula, Er is the difference between the actual and predicted P-wave transit time in the normal well section, n is the number of sampling points for the P-wave transit time curve and resistivity curve in the normal well section, and DT i Let dt be the actual value of the P-wave time difference at the i-th sampling point. i R is the predicted P-wave time difference for the i-th sampling point. i Let Er be the resistivity value at the i-th sampling point; let Er be the minimum, and solve for the values ​​of the coefficients k and m to be calculated; Based on the resistivity data of the well section requiring logging data correction and the correlation equation between resistivity and P-wave travel time, the corrected P-wave travel time curve is obtained.

9. The shear wave prediction method according to claim 3, characterized in that, The coefficients in the bivariate linear polynomial composed of the longitudinal wave time difference and density are obtained by substituting the measured transverse wave, longitudinal wave, and density data into the bivariate nonlinear function. The solution method is the Marquardt method.

Images